A clean wouldn’t hurt

These photos were taken in 2005 by MisterPX of his custom AR-15 after 15,000 rounds of Wolf ammunition. He had lubed it every 1000 rounds but not cleaned it. A thousand rounds later a bolt lug sheared off and two magazines after that a second lug was lost and at that point he decided to terminate the experiment before something disastrous happened.

MisterPX said …

Those are my pics. I took those around the 15K mark. Broke a bolt lug around 16K, second lug broke after another mag or two, then I terminated my experiment. Interestingly enough, the bolt release broke in that 15-16 range as well.

… barrel was Bushy, BCG was M1S, think upper rec. was as well. Lower was a RRA, do not recall who’s LPK.

There is a discussion about this rifle and others at ARFCOM.

[ Many thanks to jdun1911 for emailing me the link. ]

Steve Johnson

Founder and Dictator-In-Chief of TFB. A passionate gun owner, a shooting enthusiast and totally tacti-uncool. Favorite first date location: any gun range. Steve can be contacted here.


  • Clodboy

    Was this part of a scientific paper? “Production of carbon nanotubes through vapour deposition in a direct-impingement rifle” maybe?

  • Redchrome

    Interesting. I am impressed.

    Still, the fact that it *requires* so much lubrication, and shears bolt lugs is an indictment of its design.

    Since when did you ever hear of an AK/M1/M14 shearing bolt lugs? It seems downright inevitable on the AR15 (tho the high-end magneto-particle-tested bolts may be less likely to fail that way). That, or breaking the bolt in half at the cam pin hole. 7 lugs can’t be easily made to bear evenly; some are going to take more strain than others, especially when things get dirty.

    I believe LWRC (back when it was Leitner-Weise) ran one of their early piston ARs for 20K rounds without cleaning or lubrication. (Someone please correct me if I misremember, it’s been a while). Seems perfectly obvious to me that dumping hot fouling into the moving parts is not conducive to the reliable operation of anything. The AG42/Hakim was a direct-impingement system as well, but it vented directly to the open air, thus minimizing fouling buildup on the action parts.

    • Redchrome, they are simply not designed to last 16,000 rounds. It is not really a problem. I am sure someone could manufacture a bolt to last way longer at a price.

  • jaekelopterus

    If he’d been doing something as simple as lubricating with CLP you’d probably see a much cleaner firearm. Still, 15,000 is an amazing number of rounds, with or without cleaning. I wouldn’t be too surprised if an AK/M1/M14 or similar system had similar problems after such abuse.

  • Freiheit

    @Redchrome – I own a Hakim and I’d like to correct something – “…but it vented directly to the SHOOTERS FACE, thus minimizing fouling buildup on the action parts.” 😀

    The author specifically says he was using Wolf. Are there similar tests with other ammo?

    What variables existed in this test?

  • Tyson Chandler

    I am impressed also. I was thinking the same thing as Redchrome, that the results might be different with a Colt or LMT BCG that was proof fired and MP tested. I would also like to know what the barrel gas tube length were. I like mid-length gas tubes on a 16 in barrel. Could this be an instance of an AR matching an AK’s reputation for durability and low maintenance? It would be neat to see comparison test on the AK.

  • Redchrome

    I realize they aren’t designed to last 16K rounds. My point is that I believe there are better designs that very well do last 16K rounds, and do not incur a substantial weight penalty. 🙂 (When the bolt design is considered in isolation, not the gun as a whole). The XCR for instance uses an AK-style bolt with much heavier locking lugs; but still locks into a barrel extension like the AR15, for a comparable weight.

    I’m also curious what weight of buffer he was using in this gun. Whether it was a standard carbine weight, rifle weight, an H2, or something else. My experience with Wolf steel-cased ammo in semi-auto guns (even XCRs) is that it sticks pretty badly in the chamber. A heavier buffer would slow down unlocking, let pressure drop further, and thus make the thing more reliable.

    Of course, the OP didn’t state how many jams there were.

  • jdun1911

    I can assure everyone here that parts on whatever type of firearms you use will break sooner or later. It’s fact and not my opinion.

    LWRC like any other company use their test as marketing. Unless someone do an independent study I would take whatever company said with a grain of salt.

    AK parts does break. All you need to do is go to any AK forum and find out which part wear out first. For example:


  • Sian

    Great googly moogly.

    That’s impressive in several ways.

  • Jim

    I am genuinely disappointed that it failed in such a catastrophic way. The more I read and see about the AR-15 design the less I like it.

  • whizzo

    Is this also an indictment of Wolf ammunition?

  • benEzra

    “Seems perfectly obvious to me that dumping hot fouling into the moving parts is not conducive to the reliable operation of anything. The AG42/Hakim was a direct-impingement system as well, but it vented directly to the open air, thus minimizing fouling buildup on the action parts.”

    Redchrome, the AR vents to the open air as well, through the holes in the right hand side of the bolt carrier that face the ejection port. The only gas that goes into the action is residual flow that occurs after the bolt opens, and you get that with any firearm (though perhaps a bit more with a DI system). But the majority of the gas is vented out the side of the bolt carrier.

  • Tommyboy

    Wow… It looks like someone just took a s$%t in this AR and no one cleaned it. Shows what wolf does to good quality rifles.

  • Bartholomew Roberts

    Redchrome, the carbine above is using a bolt designed for the operating pressures of a 20″ rifle. In a 16″ barrel with a carbine gas system, it sees around 150% higher loads than it was designed for and still managed to make it to 16,000 rounds before a part failed.

    There are improved bolts out there that are much sturdier built; but NSWC Crane nixed the improved bolt program because all of the new designs required a new barrel extension in order to work. Using the existing barrel extension, there is only so much the design can be improved.

    Also, I don’t really feel that lubricating every 1,000 rounds “requires so much lubrication.” That seems like a pretty reasonable requirement to me since that is about 3x the amount of even the more extreme ammo loadouts.

  • Jeremiah

    It makes me almost want to cry when I see people abusing their firearms like this. It makes me sad to see someone abuse their AR-15 because there are tons of people who can’t afford one, and there are even more who wouldn’t abuse one if they had it. It ranks right up there with Glock torture tests.

    I’m not trying to be judgmental; I’m just wondering how much it’s worth to put your gun through one of these *clearly* un-scientific tests in the name of…? Internet notoriety?

  • Don

    What solder would even get close to 1000 or at a far reach 2000 rounds before having adequate time to clean the little carbon deposited? Who would take their rifle to this extreme? Why would you not maintain your rifle before it got to this extreme? This is the real world not made up forum chatter. A solder is taught that rifle is the most item/important weapon that they have and how to maintain the weapon. I also do know civilian that would get close to this extreme either. Maybe you know people who abuse their firearms. Sure pistons are nice but no one is going to run an AR without cleaning it. If they do they are idiots and make other poor decisions in life that will lead to problems bigger that a dirty AR. With regular NORMAL cleaning an AR will last for a very long time and will more than likely shoot out the barrel and develop poor accuracy way before catastrophic failure.

    Having actual test experience with firearms where real test standards were followed and seeing several firearms fail that we thought would pass with no problems. I know there is a lot to a test than a good old boy try to maintain how repeatable and how fair to all firearms subjected to these tests. Testing just one firearm doesn’t prove anything. It tells you when that firearm failed but not that everyone of similar make will fail at that round count or in the same way. As well this AR may have experienced excessive heating form firing too fast or not allowing adequate cool down. Two thing that most actual test standards allow and require.

    I doubt that outside a “test” that anyone will ever see this kind of failure.

    I’m not saying that a piston drive isn’t cleaner etc. but there are a lot parts we add to our AR’s that are not necessary for them to function for a normal life expectancy. I’ve been thinking of designing my own piston drive just to cash in on the craze.

  • Spade

    “Still, the fact that it *requires* so much lubrication, and shears bolt lugs is an indictment of its design. ”

    It was 16k rounds and the upper and bolt were M1S. M1S is one step up from the angry beavers at Century. I’d hardly claim that a M1S bolt breaking is an indictment of the design. Of course an LWRC bolt would last longer than a M1S bolt.

  • Lance

    The problem too is that he used wolf ammo which is too dirty. Stell case dont expand like brass so more carbon escapes into the action as well. The gun may have done better with american or NATO ammo.

  • Sean

    Interesting. But I have a question. WHY? Why ruin a perfectly good gun? Why waste that much ammo? Why here in Great Depression 2 spend thousands of dollars on the gun and ammo for this?

  • jdun1911

    While we are on LWRC. Here is what an LWRC rifle in 308 preform without the marketing department behind it.


  • UraniumHead

    I think the most interesting result of this is putting to rest the myth that steel-cased ammo will strain a rifle’s extractor.

  • Al T.

    Pat Rogers is running several ARs in his classes with no failures, no cleaning and lots of Slip 2000 EWL. IIRC, several are over 20k and one is over 32k. The AR platform has it’s issues (and lots of pluses), but sensitivity to carbon is not one of them.

  • MrMaigo

    Lets see an AK do that

  • Tylermar

    Wow is a piston rifle or not

  • Burst

    The AR-15 got a lot of things right. Straight line profile, lightweight low recoil ammo, (too) early use of polymer. Even the carrying handle is somewhat clever.

    But all this is to say, that direct-impingement was already obsolete in 1942. That it would be revived a generation later for a modern weapon is incredible. That the world’s most powerful army would employ it for 50 years is just insulting.

  • juan

    Very impressive
    The hk G36 did not resist more than 3000
    One indiscreet question. How much did you cost the experiment?

  • MrMaigo

    You’re missing the point, he used 3 grand of Wolf and didn’t clean it just to prove that AR15’s aren’t the jamtastic crap people call them. And the weapon isn’t even damaged, under $100 would put it back into action, and still be better than any of those ‘rebuilt’ AK’s people get.

  • Mat

    What an AR can co AK can better .Direct gas impingement is just plain stupid for a combat rifle .Reliabiliy is AK’s bigest strong point ,its gas piston design keeps things cleaner and cooler for much longer and and tolerances and spaces around the bolts working area alow for a lot of dirt before it start giving any trouble and AK locking lugs are massivey over sized.And when comparing AK’s and AR try to compare rifles developed an made at least in same dacade. It always funy when bashing the AK people end up comparing a chinese knockoff AK47 to new AR.Compare it to a Vietnam era AR ,no rails,no optic,pencil thin barrel

  • benEzra

    “I am genuinely disappointed that it failed in such a catastrophic way. The more I read and see about the AR-15 design the less I like it.”

    I’m not sure I’d call that a catastrophic failure (nothing like a “kaboom”), given that an AR’s bolt is a relatively inexpensive part that is supposed to be replaced ever 5K or 10K rounds, and when it did fail, it shed a lug but kept working, then shed another lug without blowing up the gun. If the gun weren’t being used in a torture test, the bolt would have been replaced long before this happened. There’s also the fact that this was a relatively low-end bolt, and it still held up to 16K rounds; a top tier bolt from BCM is only about $70 and would likely go longer.

    As to the amount of crud in the action, I own an AK and will say that if you run 16,000 rounds of Wolf through an AK without cleaning it, it will probably look about as dirty.

  • Ken

    What a mess. Try shooting blanks and it gets that dirty much quicker. Ahh the days in the woods playin soldier…

  • Sian

    @Mat Can you still buy that vietnam-era M16 or anything with its drawbacks? Why would you even make that comparison? The AR and AK are both mature platforms, the AK simply started that way.

    The AK47 is amazing for what it is.. a rugged lead-chucker that can be abused by poorly trained or untrained militia and reliably deliver pain downrange, but its limits are quickly reached when you put it in the hands of a rifleman.

    The post proves the rugged reliability of the AR platform, DGI or no.

  • Cameron

    @Tommyboy Errrr… I’m fairly certain that if you shot 16000 rounds of any ammunition through an AR it would look pretty terrible.

  • bryce

    Expensive experiment
    That would have cost me $4080 for just the ammo.

  • jdun1911

    Tyson Chandler,

    All MPI (Magnetic Particle Inspected) test only tells if the bolt will break in the first ten rounds or not. After the 10 round it is pretty much useless.


    You need to understand how the AR works. The bolt and barrel take the majority of the stress. Hence it is common for the locking lug or cam pin to break. Most good bolts will last over 5k rounds. The bad ones sub 1k.

    Unlike other firearms, the lower and upper receiver will last forever because the stress is taken by the bolt and barrel.


    Do you know why people abuse their firearms? So they want to know its limitation.

    You can not know how well your firearms will perform unless you go to the extreme with it.

    There are two types of people, the safe queens and the fighters.


    He didn’t ruin his AR. An AR15 bolt cost $70 and the miscellaneous parts probably cost less than $30. He probably needs to replace the barrel because the rifling is most likely running thin. Both receivers are undamaged.

    What made you think he wasted his ammo? I clean my AR between the 2k-3k range.


    You do know all those new firearms design aren’t new. They are recycle from firearm designs that dated back to the 19th century.


    It’s a DI AR.


    Wolf was $99 per 1k when he started the his experiment.


    The reliability between the AK and AR is marginal.


    I find it silly that so many people complain that you shouldn’t use wolf on AR. If your AR doesn’t take Wolf than their is something wrong with your AR.

  • MrMaigo

    Some guns don’t like some ammo, like high end 1911’s that are picky.
    Wolf is considered the dirtiest and some of the lowest powered ammo. That just furthers the argument of how wrong people are about the AR.

    I wish I had mine back then, I’d have stacks and stacks of ammo…

  • AJ

    Redchrome, of course they make parts that can hold up longer than this, but most people don’t want to spend the money on it. You can buy chromed bolt carriers that require ZERO lubrication. If you have the cash and want a rifle to run dry, anything is possible.

    JP Enterprises makes a bolt that is supposed to hold up to over 60,000 rounds. It is made from the same steel (SAE 9310) used in the transmission gears of Formula 1 race cars.Brownells sells it for $139.

  • Mark

    I personally shot next to the rifle Pat Rogers has his students shoot in class that has over 35k through it. Not for noting does it have the Filthy Fourteen nickname. Internally it looks much like the picture in this post does (a lot more gooey, if you can picture that), but it continues to run like a sewing machine. You get a bit dirty shooting it, trust me.
    They key is good lube used liberally and proper maintenance intervals. Extractor springs, bolts, carriers, all wear out, and need to be replaced at intervals.

  • Vak

    Okay, so if anyone could send me three grands worth of wolf 7.62×39 and a AK, I would be glad to do the testing for them. (and keep the AK to myself afterwards)

  • Redchrome

    Lots of comments to respond to. Thanks everyone for being polite. 🙂

    * I continue to be impressed by how reliable AR15s are being made to be. They really don’t jam up once every few magazines as I had once believed.

    * I’ve read Pat Rogers’ articles and I believe he points out that bolts are designed to be replaced before 10K rounds. Also that they can withstand a lot of fouling as long as they have enough lubrication (sort of like adding fiber to your diet!)

    * My point is not that the AR doesn’t work; but merely that there are better designs out there. Since when do you replace the bolt on a Mauser or an M14 or an AK at 10K rounds? You shouldn’t need to make the bolt (or any other part *cough*magazine*cough*) out of Expensivetanium to have it work reliably; a different design will accomplish the same thing for less money.

    * benEzra — the ‘majority’ of gas may be vented to the air, but there’s still plenty of fouling dumped where it doesn’t need to be.

    * Bartholomew Roberts — yep. In the 20″ bbl the design was marginal. Going to a 14.5″ barrel made it worse because of the higher pressures. Going to a 16″ bbl but keeping the gas system intended for the 14.5″ bbl made things worse yet because of the longer dwell time at high pressures. 16″ bbls should have midlength gas systems; tho it’s really all a tradeoff. You can put a “rifle-length” (intended for 20″ bbl) gas system on a 16″ bbl and it will still work; but it’ll be more picky about what ammo it works reliably with.

    I was not aware that Crane had come up with a better design. Thanks for informing me. Backwards compatibility is a killer. 🙁

    * UraniumHead — the article didn’t say how many failures to extract or failures to feed he had; nor was there a side-by-side comparison done with M193 (brass-cased) ammo. Shoot some steel-cased .45 ACP ammo in a revolver with moon clips and you will likely feel a substantial difference between that and brass-cased ammo.

    * jdun1911 — yep, I know the AR bears most of the strain of firing in the bolt locking to the barrel extension. It’s a great design that saves weight. The problem is that they went overboard on saving weight, at the cost of reliability. Beefier locking lugs would cure much of the problem.

    Just because it can be made to work doesn’t mean it isn’t broken by design. A great design doesn’t need special materials and special conditions to run well.

    Still I continue to be impressed by how far the AR design is being kludged along, and how reliable people are demonstrating it to be.

  • puppezed

    Any info of the failure rate during this test?

  • I shot one of Pat Rodgers’ ARs (a DPMS, if memory serves) in his carbine class last year, and, although the gun is technically operative, I referred to it not as my rifle but as my malfunction-clearing practice platform. Damn thing would malf at least once every two mags, no matter which mags I used.

  • Tony Chow

    The DI system is instrumental to making ARs the most accurate semi-automatic rifle ever made. It’s why ARs so swiftly displaced M14s in High Power competitions as soon as the necessary improvements were made to the platform, and it’s why you won’t see any of the new crop of piston guns being used by competitors any time soon.

  • Nick

    Some food for thought. These were filmed here in AZ, and make a pretty strong statement for the differences in reliability between the AR and AK platform.

    AK guys, be ready to stomp, shout, curse, and defend your “jam-free” rifle, LOL.



  • Redchrome

    when I saw those videos a while ago it was one of the things that changed my opinion of the AR’s baseline reliability.

    However, coupled with these videos:

    It is fairly obvious that if you put dirt into *any* gun you’re going to get some issues.

    Reliability is not a binary (works always/never works); but instead is a continuum (works almost always under these conditions/most of the time under these/ rarely under these). Certain features increase reliability. The AR has some. The AK has others. No one has yet put together all the best features into one rifle. 🙂

    (And I’m still taking my AK if I want a semi-auto gun that I trust will go bang when I want it to).

  • What solder would even get close to 1000 or at a far reach 2000 rounds before having adequate time to clean the little carbon deposited? Who would take their rifle to this extreme? Why would you not maintain your rifle before it got to this extreme?

    One word — Bastogne.

    No one would willingly take their rifle to this extreme in combat, but that’s the thing about combat — the enemy forces you to do all sorts of things that you wouldn’t normally do, usually towards the goal of forcing you to die when you don’t want to.

  • KiwiBayly

    Looks like a similar test I saw on this standard FAL http://www.falfiles.com/forums/showthread.php?s=&threadid=68486

  • jdun1911


    Go ahead and redesign the locking lug.


    I’m pretty sure there are a number of cases where rifles needs to run over few thousands of rounds before cleaning in Iraq and Afghanistan. It’s not like major engagements last for a few minutes. It might runs for days or weeks.

  • zack

    I would still never, ever use wolf garbage in my weapons.

    • I won’t use corrosive ammunition even in my Warsaw surplus guns, but other than that I don’t care.

  • DJ

    One thing DI gives you is a lighter weapon. It’s hard to find DI and Piston ARs with a similar enough barrel profile to compare weights directly, but my estimates put the weight penalty for a Piston gun between a half-pound and a full pound. In a light 5.56mm/.223 carbine, that’s a 10 to 20 percent increase that you’ll have to hump around EVERY MINUTE OF EVERY DAY, for a supposed increase in reliability that most folks will never see or need.

    I have a pre-ban pencil-barrel Colt carbine that is a joy to carry and shoot, and is over 2000 rounds of miscellaneous brass-case ammo without a cleaning right now (lube is added periodically, as it looks to be needed). No stoppages since the informal experiment began.

  • Nadnerbus

    I’ve had Wolf ammo jam up my (pre redesigned) mini 14 after only a few hundred rounds. When I feed it brass cased anything, it just keeps chugging no matter how many rounds. Wolf has a reputation for being dirty for a good reason. It’s also harder on the chamber, and harder on the extractor because the bi-metal case is not as soft as brass.

    The AR is also a get what you pay for rifle. From what I have read over the years, DPMS and Bushmasters tend to be towards the bottom in quality and reliablity (though still pretty good for civilian shooting), Rock Rivers towards the middle to top, and then LMTs, Noveskes, and LWRC at the top. Everything else fist somewhere in that continuum. Any brand can have lemons, but I have put several thousand rounds though my Rock River A4 (Lower is California off list, everything else is RR) and the three our four jams I had came from one bad John Massen magazine. Of course I fed it all decent quality brass cased ammo, and cleaned it after no more then six to eight hundred rounds.

    I just can’t be a AR hater. It’s a good design, no matter what some people think. It just has it’s limits, like any other rifle. My vaunted Springfield M1A, that cost me twice what an AR would at the time, jams constantly, even with a “free” lifetime service repair. My DPMS 308 AR worked right out of the box, with great accuracy, and not a single FTF or FTE.

  • Makes me wonder, how bad could it be to shoot black powder .223 loads?

    • Cemetery: I just saw your post on blackpowder .45. I don’t think it would be a good idea to shoot blackpowder .223 from a AR. The pressure curve would be all wrong and there would probably be problems cycling.

  • Allan Spets

    Look at the mfg of the bolt. Model one sales is bottom tier company. None of the parts used were mil-spec.

  • TheGimp

    I’m just wondering if the extra crud didn’t add to the internal pressure thereby causing the rifle to fail earlier than it should have.